CtrlK
BlogDocsLog inGet started
Tessl Logo

agent-desktop

Use the built-in `Computer` sub-agent with `agent-desktop` for macOS desktop automation. Apply when a task needs application launching, accessibility snapshots, stable element refs, window focusing, semantic clicks/typing, or visual confirmation outside the browser sandbox.

75

Quality

92%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

85%Weight 40%Scale 1-3

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-structured, actionable skill body with a clear workflow and safety checks. Its only weakness is mild redundancy across sections that could be consolidated.

Suggestions

Consolidate the repeated guidance about ref staleness and preferring refs over coordinates — it appears in Requirements, Tool guidance, and Reliability rules; state each rule once in the most relevant section.

Trim the 'When to use it' bullets, which restate triggers already present in the frontmatter description, to avoid duplication.

Merge Preferred flow step 5 ('After any UI transition, snapshot again before reusing refs') with the 'snapshot -> act -> snapshot loop' reliability rule into one authoritative loop description.

DimensionReasoningScore

Conciseness

The body is largely lean and assumes Claude's competence, but it restates ref staleness and 'prefer refs over coordinates' across the Requirements, Tool guidance, and Reliability rules sections, which could be tightened to a single statement.

2 / 3

Actionability

It gives concrete, per-tool guidance with specific usage rules ('prefer interactive_only', 'use ref values from the latest snapshot', 'pass an element ref, not raw coordinates'), which is highly actionable for an instruction-only skill.

3 / 3

Workflow Clarity

The numbered Preferred flow has an explicit re-snapshot checkpoint, the Reliability rules encode a snapshot->act->snapshot loop, and the Blockers section provides safety validation for destructive actions.

3 / 3

Progressive Disclosure

With no bundle files present, the content is well-organized into clearly labeled sections (When to use, Requirements, Preferred flow, Tool guidance, Reliability rules, Blockers) appropriate for a simple single-purpose skill.

3 / 3

Total

11

/

12

Passed

Description

100%Weight 40%Scale 1-3

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, specific description with an explicit trigger clause and a clear, distinct niche. It assumes third-person imperative voice and avoids fluff or over-claims.

DimensionReasoningScore

Specificity

The description lists multiple concrete actions — 'application launching, accessibility snapshots, stable element refs, window focusing, semantic clicks/typing, or visual confirmation' — rather than vague language.

3 / 3

Completeness

It answers both what ('macOS desktop automation' via the Computer sub-agent with agent-desktop) and when via an explicit 'Apply when a task needs...' trigger clause.

3 / 3

Trigger Term Quality

It covers natural terms a user would say ('application launching', 'clicks/typing', 'window focusing', 'visual confirmation') alongside the technical vocabulary, giving good trigger coverage.

3 / 3

Distinctiveness Conflict Risk

The niche is clear and narrow — host macOS desktop automation 'outside the browser sandbox' — making it unlikely to fire for browser, shell, or repo-file skills.

3 / 3

Total

12

/

12

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
superagent-ai/grok-cli
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.